Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner
National Hospital Available Beds for Emergencies and Disasters (HAvBED) System

Public Health Emergency Preparedness

This resource was part of AHRQ's Public Health Emergency Preparedness (PHEP) program, which was discontinued on June 30, 2011, in a realignment of Federal efforts.

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Appendix B. HAvBED Advisory Group Meetings

Meeting Minutes
Thursday, July 15, 2004

Secretary's Conference Room
Office of the Assistant Secretary for Public Health Emergency Preparedness
Hubert Humphrey Bldg., 200 Independence Ave., Washington, D.C. 20201


Lt. Col. James Baxter
Rhonda Earls, M.S.N., C.N.M.
Matthew Payne, M.P.A.
Robert Blitzer
Capt. Ann Knebel, R.N., D.N.Sc. 
Sally Phillips, R.N., Ph.D.
LCDR Sumner Bossler
Lt. Col. William Kormos
Peter T. Pons, M.D.
Shayne Brannman, M.S., M.A.    
Sheri Eisert, Ph.D.
Dean Ross
Stephen Cantrill, M.D.
Capt. F. Christy Music, M.Sc., U.S.N.
Kevin Yeskey, M.D.
Robert Claypool, M.D.
Michael Nugent

Office of the Assistant Secretary for Public Health Emergency Preparedness (OASPHEP) Considerations

Mr. Blitzer

Attributes Mr. Blitzer would like this effort to include are:

  1. Utilize existing systems. Capitalize on existing systems that are currently in use and working. Should not be intrusive.
  2. Inter-operability. Should be interoperable with existing systems such as HARTS and the National Disaster Medical System (NDMS).
  3. Adaptable. Data should be made available by Region and have the ability to reconfigure information to meet the event at hand.
  4. Real Time. Timely, accurate data should be available within 2-4 hours of an event.
  5. Bed Definition. Not just bed count; need number of available staffed beds, specialty beds (i.e., burn, neonatal, coronary care unit [CCU], intensive care unit [ICU), ventilators while not being intrusive.  Want all beds to be accounted for, including public, private, Veterans Health Administration (VA) and Department of Defense (DOD) beds. Divide into pre-hospital assets, hospital assets and alternative care sites.  Build on HARTs.
  6. Utility to the Department of Health & Human Services (HHS). This system needs to help HHS make sound and timely decisions on how the Federal government can best help local and State resources respond to an emergency, based on verifiable sound data. This data will drive policy decisions like redirecting resources (staff, equipment, supplies) to a region in need from other regions, pre-positioned resources, NDMS, etc. Data from this system could drive aero-medical evacuations from one region to another.
Mr. Ross

System needs to be dynamic, bi-directional, scalable, relational and spatial. Attributes to address include: 

  1. Legislation. Mr. Ross mentioned that the federal government is currently prohibited from collecting "routine data" from local agencies.  This prohibition is outlined in existing OMB legislation.  The question was posed as to how NDMS is allowed to receive data from locals.  Two conditions allow NDMS to collect such data:
    1. The NDMS has a signed memorandum of understanding with each participating hospital giving permission for such collection.
    2. The NDMS collection of data is not considered "routine" (We were not provided a definition of what OMB defines as routine). 
    Dean also mentioned that he has had several conversations with hospital executives in which he discussed their willingness to provide bed availability data.  He Stated that they seemed to be resistant because of competitive concerns.
  2.  Scalability. All meeting participants agreed that the HAvBED system must be scalable.  Hospitals should be able to determine bed availability in close proximity to their institution and move out from there as capacities are exceeded. 
  3. "Disaster EGOV" Compliance. HHS and Department of Homeland Security (DHS) initiative; compliance will be developing uniform data standards. 
  4. Geospatial Elements. Mr. Ross commented that geographic information systems (GIS) and geospatial analysis capabilities should be an essential component of the HAvBED system.
  5. Erroneous Inputs. During the design phase of HARTS, Mr. Ross visited several of the existing bed tracking systems including the Hospital Emergency Response Data System (HERDS) and the Washington system.  During these visits, he encountered instances in which erroneous data had been entered into a system.  Need to identify by analyzing variation in trends and putting reasonable limits using capacity numbers.  This could be an objective of the HAvBED exercise, to validate via site visit or internal confirmation that data entered has been accurately entered. Dr. Phillips also suggested validating the numbers through the testing component of this task order.
  6. Scope of Hospitals.
  7. Data Dictionaries.
  8. Diversity of Data Locations. Mr. Ross described the HHS' "Hug the Box" policy.  In order for HHS to integrate a technology system into their operations they must own and operate the server.  Existing systems used by HHS contain redundant and secure data centers located in 5 different sites around the country.  In the next two weeks, Office of Management and Budget (OMB) legislation will be approved requiring federal systems to demonstrate this type of "diversity."
  9. Cannot be totally Web dependent. Both Dr. Cantrill and Mr. Ross cautioned the group about our increased dependence on Web-based and IP systems and tools.  This infrastructure could be damaged during a natural or manmade disaster; Mr. Ross suggested the HAvBED system should consider telephonic inputs.  Data entry considerations include the use of telephone data entry using automated prompts and the use of stand alone alert equipment in EDs that prompts hospitals to enter data. Ideally will have single device system-no email or phone calls. 
  10. Data-Field Flexibility. Must include flexibility to add or delete fields.
  11. Permissions. Who should have access to what?

Definitions: Beds

Available Beds: Licensed, Staffed and Equipped

Using number of licensed beds in defining capacity is dangerous; hospitals may have entire wings shut down and still have those beds on record.  DAS Blitzer prefers to have beds distinguished by staffed beds and other non-staffed beds.

Dr. Yeskey is concerned that bed definition/categories do not reflect the services needed (i.e., food, isolation etc). The definition needs to reflect more than just staffed and licensed.   It is important to understand service capability or what type of care can be provided and how the beds will be used. The HAvBED system may not be able to define standardized definitions for bed classifications. Ideally, the tool would enable a facility to respond to services needed. Defining beds based on level of care needed by the patient will help limit the variance in reporting definitions amongst clinicians and facilities.  A ventilated bed category might be more explicit than a critical care category such as NDMS uses, again terminology should reflect patient need.

Capt. Music wants to follow patients through the continuum of care and therefore prefers categories that include rehabilitation, nursing home, etc.  She also feels the categories should include civilian and Federal (DOD, VA, military).

The HAvBED system should include what an institution's adjusted bed capacity could be if augmentation was provided.  E.g. Staff, equipment and external support.

Current NDMS Categories

Mr. Ross mentioned the need to look at psych beds since there is a need based on expected hysteria of population. Dr. Yeskey does not believe that lumping medical and surgical beds is a good idea and they should be separated.  There is a need to add isolation beds. 

Dr. Yeskey also thought it was important to include a hospital evacuation category so that we know when a bed is about to be available.  Civilian clinicians need to think about the concept of evacuation upon entry to the medical facility.  In a military context, the term evacuation is used to describe patient movement out of a medical facility to another environment such as specialty services or rehabilitation.  Dr. Yeskey believes that civilian clinicians currently consider this concept upon discharging patients; however given a rapid influx of acutely ill or injured patients some decisions will have to be made early as to whether the hospital can provide the continuum of care required by the patient.  Dr. Cantrill questioned whether civilian health care providers could make such decisions.

Comments were presented regarding medical disaster management, specifically as to whether it is more effective to evacuate patients in a disaster area or to deploy medical resources to the affected area.  In the cases of specialty care, it may be more beneficial to move these patients to a site with this capability.  However, in a large-scale medical disaster, patient movement out of the affected area is too resource intensive.  There was consensus amongst the participants that the NDMS patient evacuation resources would be quickly diminished in a medical disaster scenario.

There seemed to be a consensus among the Advisory Board that the project should not limit its data elements to those currently used by NDMS and that these are not meeting the HHS regions' needs.  This is another reason why it is important to look at the others systems that have been developed, since other people have been looking at these same issues for a while. 

HAvBED Data Elements

LCDR Bossler stressed the need for simplicity and proposed focus on a few categories of beds.  This project should review the other systems to see what they collect and interface with what they currently collect to the best of our ability. Most of the project effort should be spent on creating data and reporting uniformity among the already existing system. Then we could develop categories/definitions and data elements that would be most appropriate and provide them in the form of recommendations in the final report. Adding additional data elements could decrease the utility of the HAvBED concept. 

Dr. Yeskey thought we should develop the ideal list of categories and augment the systems to the best of our ability to capture this data, particularly since this is a national system. This project should identify an all-encompassing, best-practice data set that would be used in the ideal bed tracking system. It would be disappointing if the threat of "pushback" from private industry vendors or current system participants limited identification of these data elements. These ideal elements should be based on federal, State and local needs.

Total Number of Data Elements

Originally, the NDMS tracked 16 different bed categories; this number was decreased to 6 and increased reporting compliance. HARTS collects 10 different bed categories including the number of isolation rooms available.  TRAC2ES collects bed information based on snapshot. 


Some discussion on defining capacity as what you can do with your current resources with minimal augmentation.

Definitions: Time

"Real-Time": Data Input Frequency

"Real time" data entry and display was not really defined in this meeting.  Most in the meeting agreed that a snapshot of data such as that collected by NDMS is insufficient and limited.  Bed and resource availability is too dynamic. 

Comments were made about the frequency of data input in an ideal system.  HARTS data can be updated every four hours.   The question was posed regarding whether day to day data input is necessary; instead, could an effective bed tracking system only include data entry during a large-scale emergency?  The group favored the ability to obtain data at a variable rate based on the type and scale of an event.  The system would have to be extremely simple and intuitive for a user.

HHS has developed system that uses satellite band for medical warnings like the weather service.

Maximum: Available in 72 Hours

The NDMS system provides a data collection field called "maximum beds."   This is defined as the number of potentially available beds within a 72 hours time period.  Most users agree that this number is a "best guess," but Dr. Claypool believes that the current threat climate necessitates some similar attribute in the HAvBED system.  The existing snapshots of bed availability provided by many of the systems are helpful, however prolonged incidents spanning weeks or months will require projections of bed availability.

Patient Tracking

Capt. Music emphasized the need for patient tracking.  Capt Knebel cautioned with the need to either focus on the tracking of assets or the tracking of patients.  Lt Col Baxter explained that TRAC2ES tracks the patient while in transit.  Once they are discharged, other systems capture the data.  As part of this task order, Dr. Phillips explained that patient tracking can be part of the asset bed tracking component such as developing the fields, while not necessarily collecting actual data.  FEMA is expected to have 10 emergency coordinators among the 10 regions to assist with patient movement.

The only existing system that seems to track from time of injury to hospital destination is the St. Louis system with involves a cooperative relationship between Raytheon and EMsystems.  Patients are bar coded for patient tracking and destination is provided using EMsystems. 

The Advisory Board approved of the concept of when moving patients, priority should be to move them to local area hospitals, even if this requires minimal hospital augmentation.  It would be preferable to move the assets to where the patients are located rather than move the patient to the part of the country where there are assets.  This is particularly applicable to a communicable event.

HARTS System Overview

Mr. Ross presented an overview of the HARTS system and gave a handout that Dr. Eisert collected.  He described HARTS as a comprehensive system that tracks numerous different attributes including blood, equipment, etc.  Can add fields, GIS component, has event log that can be used by participating hospitals.  Once hospitals are "opened up" (those medical centers that are affected by an event) they are given an opportunity to enter data.  If hospital is a hub for both blood distribution and medical care, the facility can have access to both HARTS components.  HARTS has a chat function and derives its facility data from the American Hospital Association.  During events, HARTS can specify the frequency in which users must enter data and then track hospital compliance via colorimetric scales such as red, yellow, and green. Looking at TRAC2ES system to incorporate patient movement.  HARTS activates hospitals and other medical facilities by placing a phone call and deploying emergency response teams to the affected area. HARTS can be updated every four hours.  HARTS has admin section, hospital section and blood section.

National Health Alert and Warning System

Mr. Ross provided a brief description of the National Health Alert and Warning System.  This non-terrestrial system is currently in testing and production; in summary, it includes an inexpensive alert device that can be placed in a public health center or hospital and has the capability to transmit voice, text, and data when activated by HHS.  Staff will receive an audible tone to alert them of an incoming message.

Choice of Tracking Systems To Review

  • Illinois—HHAN system.
  • St. Louis—hybrid system with Raytheon patient tracking function and EMSystems patient destination.
  • FREDs Pennsylvania State?
  • California—Capt Music refers to Jeff Rubin (Reddinet?).
  • North Carolina?

Additional Participants

  • DHS/NDMS—Contact Bob Jevec.
  • American Red Cross—should have representative although Capt. Knebel Stated that Dr. Claypool could represent this group.
  • American Hospital Association representative.

Next Meeting Date

The group agreed on the week of October 11. HAvBED project team is looking at October 14 or 15 for the next meeting and will solicit votes.

HAvBED Advisory Group Meeting Minutes
Thursday, October 14, 2004

Deputy Secretary's Conference Room
Office of the Assistant Secretary for Public Health Emergency Preparedness
Hubert Humphrey Building, 200 Independence Ave.
Washington, D.C.  20201


Lt. Col. James Baxter
Ivan Gotham
Sally Phillips, R.N., Ph.D.
LCDR Sumner Bossler
Nathaniel Hupert, M.D., M.P.H.
Peter Pons, M.D.
Shayne Brannman, M.S., M.A.
Jeffrey Lowell, M.D.
Dean Ross
Stephen Cantrill, M.D.
Duane Mariotti
Daniel Salazar
Robert Claypool, M.D.
Capt. F. Christy Music, M.Sc., U.S.N.
Maj. Wayne Surratt
Rhonda Earls, MSN, CNM
Dawn Myscofski, Ph.D.
Carrie Vinci, Sc.M.
Sheri Eisert, Ph.D.
Andrew Nunemaker
Elaine Wolff, M.H.S.
Michael Feeser

Hospital Beds Definition

Staffed and Unstaffed Beds

The project team proposed that unstaffed bed count is a useful data element reflecting an open bed that a patient could be regulated to if staff were supplied as an asset. The advisory group felt that contractual pools of staff may allow quick staffing of unstaffed beds. The question becomes whether these beds are "staffed" or not and what is the time frame? Other members noted that many new staff may not be familiar with the hospital equipment and procedures and need orientation.

Vacant Beds

The term "vacant" beds may not be as intuitive to users as "available" beds used by NDMS. Most hospitals are familiar with the term "available," but may not understand "vacant." 

Data Elements: Current Capacity

We must collect information that is actionable, not just collect for collecting sake. This project needs to separate things that are "nice" to know versus things that are "need" to know.   This project should also beware of mission creep and this first look through should be concise as to what we "need" to know.

This system should contain the bare essentials but other data besides bed count is crucial. Dr. Claypool is very interested in other data elements and suggests tailoring data collection to nature of events. Beginning with bed availability for simplicity sake to demonstrate the proof of concept may lead to a more refined dataset once the feasibility has been proven. This project should also be aware of looking at things at an academic level but not necessarily paying attention to what assets could actually be moved based on this information.  This project needs to address the daily trip wires and attributes.

Bed Types

If the hospitals are given no definitions they can make different beds into different types and the counting is inconsistent. Suggestions include making sure burn beds are ABA certified. ED diversion status is a more useful data element than number of ED beds since these beds can be turned over quickly.

Ventilators and ICU Beds

The Washington system collects ventilator status but not blood supply information. Ventilator status is located on the resource page and every two months requests a total number of ventilators from each hospital.   Most of these ventilators are attached to a patient and are not available. This project may want to differentiate between "vacant" and "used" ventilators. 

A concern from the last meeting referred to throughput and that additional resources beyond beds will be necessary to count. Counting ventilators addresses the issue of bed functionality because utilizing ventilators available and ICU bed count provides the necessary information for encompassing services needed for critical care. If other resource data elements are included this will address the issue of noting bed service capability. 

Since there is a lack of respiratory therapists, some suggested including the staff in the ventilator definition.  Another note of caution arose in that an ICU bed count may identify ICU beds that meet all other ancillary needs such as nurses, but do not include a ventilator. Implicit in that idea is that every ICU has to have the capability to do everything to every patient, which does not hold true for most hospitals.  A new frontier is going to be using ventilators in non-traditional settings.  Oxygen is also important to address.

Operating Rooms

This project should look at defining what an available OR is (i.e., canceling elective surgeries) including requiring staffing with a surgeon, anesthesiologist, and nurses.  The reporting of available operating rooms may be more meaningful in binary form such as "capable" and "not capable."

Federal Assets

Federal beds are a community resource. There are DOD memorandums of understanding that will provide mutual support. The VA will be told that they can make beds available to the public. DOD and VA resources should be listed but flagged until authorized to participate and accept civilian casualties.  NDMS and Tricare hospitals should be tagged in this system as well. 

For civilian facilities, there are no formal agreements and sharing of assets will be up to the locals. With NDMS, the DOD coordinates the patient transportation component. The largest issues arise from reimbursement.

Data Elements: Surge Capacity

The question was posed: should this system forecast data or is it beyond the scope of this project? The committee agrees this project should use NDMS bed availability type surge counts, instead of the proposed "unstaffed" beds category. When discussing surge, we should address not only what will become available but also what you may need in the 1-3 day range. There are two types of surge capacity and capability that should be counted: Hospital and Alternative Care Facility (ACF).

Hospital Bed Capacity

There are many different suggested hourly projections necessary for planning and responding to a mass casualty incident (MCI) such as 8-, 24-, 48-, 72-hour estimates. Each different stakeholder requires a different length projection and this group agrees that a lowest common denominator minimizes the burden on the hospital, for example 12-18 hour shifts.

NDMS surge bed counts (or deliberate planning beds) are requested at minimum (24 hour) and maximum (72 hour) projections, which are a guess for the civilian hospitals reporting.  The 24-hour estimate becomes a 48-hour projection when done every 24 hours.

Local and regional planners rely on and value the smaller hour projection where as multi-State regional and federal planners find the longer projections helpful.  The focus of your view is based on who you are and what you need. The hospitals may not know what they need and DHS will just start sending assets. The hospitals need to get better at knowing how many of what they need. They also need to think about sustainment. The 24-hour estimate becomes what from a planning perspective could be asked for in a 72-hour window.  

Providing a 24-hour NDMS type bed capacity projection will be very useful in the buffer zone of the event, but not the cross hairs. The incident will move out to overwhelm the next hospital. It provides an increased depth of information to the areas that haven't been affected. State or county managers could also use this information to decide where to place casualties in the next 24 hours.  The 72-hour projection gives a regional idea of where to place casualties.

The Washington System notes that multiple hour projections are too intuitively difficult and cumbersome for hospitals to provide before an incident occurs.  From their experience, they feel most hospitals can and would be willing to supply a 24-hour estimate. These estimates should be at minimum provided daily, during an event, every 24 hours. 

In conclusion, the board advised a 24-hour forecast with a definition if you cancel elective surgeries and discharge patients what is best guess of how many beds could be opened in 24 hours. The project will need to address in detail defining hospital surge capacity beds in a standardized way. For example, cancel elective surgeries and discharging patients. For a chemical, biological, radiological, nuclear, or explosive (CBRNE) incident, there is a need to decide the requirements for acceptable standards of care. The question that the hospitals need to ask themselves is: What do we have over and above what we are running right now?  What could we make available to accept an active case? Surge capacity requires a common understandable definition; potential capacity can change every ten minutes and depends on how frequently the polling occurs. The number of operating rooms and recovery beds will change with every operation therefore a binary capability may be more relevant.

Alternate Care Facility (ACF) Bed Capacity

The project should contain a placeholder for ACFs so that ACF bed counts do not co-mingle with the current hospital capacity information. These surge beds off-site could be entered early if known, before an incident, minimizing data entry burden and assisting in planning. For this system, a bimonthly or quarterly question of what could be made available may suffice. This information should be shared locally, to address how quickly you could move the patient between facilities locally. This would also require local memorandums of understanding and emergency medical services (EMS) integration. In addition, this project could graph capability to project needs and capabilities out or graph availability of beds over time to follow trends.

ACF bed count should be built in to this system, but would require ownership of this feature.  The committee agrees there are two main organizational levels that deal with ACFs and could report capabilities: hospitals and State (regional) planners. Hospitals could track and count according to their mandated Joint Commission (JCAHO) hospital plan for an alternative care site (ACS). This is the planning phase number associated with each hospital ACS or planned capacity.

Alternatively, States have been forced to look at the regionalization of off-site care and have allocated certain areas as an ACF. These are pre-designated in areas like New Hampshire and New York stadiums according to The Concept of Operations for the Acute Care Center, U.S. Army Soldier and Biological Chemical Command (SBCCOM), 2003. Colorado has equipment but has not pre-designated ACFs yet. In order to obtain this type of planning data, the State commissioners may be. In addition, most regions have benchmarks. The regional plans are based on certain projections such as the ability to handle 500 victims per million population or to accommodate a 20% surge.  This information could aid in estimating what the ACFs in area can provide.

EMSystem has already added ACFs such as shelters during incidents like the hurricanes in Florida.  With this feature they can track availability of shelters for DHS, HHS and the Red Cross.

In conclusion, hospitals should enter 24-hour projections daily and an ACS planning projection quarterly.  Regional/State planning information should be entered for an area's ACF planning projection quarterly. The data enterer is still unidentified and may be a long-term goal.

Systems Interface

The interface will be with each distributed hospital resource tracking system, not centralized. For the regions of the country that are not covered by each of the existing systems, the project will develop an individual hospital interface.

The Health Resources and Services Administration (HRSA) suggested that operational test cases for the individual hospital interface represent a diverse group of health facilities across the country. For example, these could include a health care system in a mid-level city such as Intermountain healthcare or a smaller metropolitan area that does not utilize one of the existing systems.

Crossing National Borders

Regarding crossing borders and sharing resources with Canada, Detroit and Ontario have been doing this for many years.  3,000 nurses move every week from Canada to the U.S. (? or vice versa) for better pay and have apartments in Detroit. They work for 3 days and then cross the border back to Canada. One third of nurses are "border crossers" in that area.  We lose care providers this way. Sharing of resources with Canada has already begun. The ministry of health runs the Ontario bed tracking system and Dr. Claypool may be able to provide a link to the person he has been working with related to cross border issues.  Dr. Claypool suggests pursuing the Ontario system. LCDR Bossler questions if the Ontario system is a provincial system vs. Health Canada.


Mr. Ross mentioned 50% of hospitals had no data capability after the earthquakes in Japan.  The network and internet NCS experienced T1 shutdown. Redundancy issues, although a valid concern, will not be addressed in this study.

Data Collection: Routine and Incident Dependent

The data should be able to communicate "I am becoming saturated, these are my needs." The Federal government cannot poll the hospitals regularly. This project should not only collect what hospitals are normally gathering and reporting, but should be able to be expanded during an event to address incident dependent data collection. EMSystem® collects this information as an event. HERDS activates certain data fields based on the event type. Routine data must be separated from an open and closed event, not a day-to-day usage. This project may want to think about scalable data entry based on nature of event and if there is an event.  

HARTS System

Mr. Ross brought up HARTS which collects real time data during an incident. Each hospital has been given a log in and a password. Illinois has a similar system. A portal sends information to HARTS and sub populates additional fields. This system will port data one way but may want to look at porting data the other way. HARTS was created Federally and is not proprietary.

HARTS has been used real time, for example during the hurricanes in Florida. The project team requested an offline discussion of exactly which data has been collected at what time.  The granularity of the data is of concern. For example, during the hurricanes HARTS collected information on hospital structural damage, power sources either commercial or generator and information was sent to the States and power companies. 

System Users 

The end-users of this system need to be clearly defined. It is important to look at how this system will be used and in what time frame.  Different levels of planning and response require different data sets. The required datasets vary for emergency departments (EDs), regional and State needs.  A concept of operations for this product should also need to be developed for all the anticipated users. This type of operational discussion is larger than the technical discussion. This system is collecting data for the local EMS responders all the way to Federal planners.

Some of the issues regarding the users of this system include operational versus planning usage. Responders utilize different information than planners. In addition, the complexity of each of these users is also an issue. Both operational and planning users may service the local, regional or national level. There are different data elements and time sensitivities required for local or regional planners and responders.  Operational users may use this tool to aid in regulating staff from another community to an event or to regulate patients from an event to another area.

The staff movement concept of sending staffing resources to an event, instead of moving patients away, was proposed to aid in containment for a biological incident. Eric Tolbert of DHS is working on some of these issues including volunteer credentialing.

The Trac2es team noted that they have a synopsis of patient regulating procedures as part of the Global Patient Movement Requirement Center (GPMRC) that they will review with the project team during the site visit.

In terms of usage incentives, this project may want to provide links to other disaster response information and systems such as HAN. This tie adds every day value to have extra information available for users. Common practices augment information and minimize additional data entry. You can ask people to populate the fields but they won't do it unless it benefits them.

Next Meeting Date

The next HAvBED Advisory Group meeting will be January 27, 2004, in Washington, DC.

HAvBED Advisory Group Meeting Minutes
Thursday, January 27, 2005

Deputy Secretary's Conference Room
Office of the Assistant Secretary for Public Health Emergency Preparedness
Hubert Humphrey Building, 200 Independence Ave.
Washington, D.C.  20201


Lt. Col. James Baxter
Michael Feeser
Sally Phillips, R.N., Ph.D.
LCDR Sumner Bossler
Nathaniel Hupert, M.D., M.P.H.
Peter Pons, M.D.
Shayne Brannman, M.S., M.A.
Jeffrey Lowell, M.D.
Col. Anthony Rizzo, M.C., S.F.S.    
Stephen Cantrill, M.D.
Duane Mariotti
Daniel Salazar
Robert Claypool, M.D.  
Capt. F. Christy Music, M.Sc., U.S.N.
Christopher Felton, M.D.
Rhonda Earls, M.S.N., C.N.M.
Andrew Nunemaker
Jacob Dye, M.S.
Sheri Eisert, Ph.D.
Kevin Yeskey, M.D.
Padmini Jagadish
Deborah Levy, Ph.D.


Dr. Cantrill

AHRQ Bioterrorism Research Overview

Sally Phillips

  • Project interface:
    • Surge capacity requirements for tracking.
    • Mass causality movement of patients.
    • Strategic national stockpile.
    • Cross training.
    • Interface with other projects like HAvBED.

Software Specifications Summary/Definitions/Screen Shots—Frames

Dr. Cantrill

  • ABA Certified is limited, not including the 43 ABA verified, there are 132 ABA certified hospitals nationwide.  Change to "Burn Unit Bed in a Burn Center" or "Burn ICU beds in Burn Centers"
  • Overstating capabilities.
  • If not ABA certified, pulling from same certified critical care nurses for ICU beds.
  • Add comment that staff for beds should not draw each other individual hospitals.
  • Negative air flow?
  • Beds must be mutually exclusive.
  • Need to have a dialog with ABA to potentially expand the burn bed data field.

Dr. Claypool

  • What is it meant by "currently not in use"?
  • Change to "can be made available in 2 hours"

Col. Rizzo

  • Virtually every operating room available for immediate receipt of patients.
  • Change to  "available for prompt receival of patients." or "Staffed ORs"
  • Ventilators—total #
  • Delete ventilators—"not in use"
  • Change "Decontamination" to "Mass Decontamination"
  • Emergent Department Status—change to:
    • Open.
    • Closed.
    • N/A.

EDXL Group Discussions

  • XML schema EDXL.
  • XML: Data tagged and organized allowing the receiving system to know where to place the data.  Snapshot of the data available and how often does it change.

Presidential Inauguration Bed Tracking Report

Lt. Colonel Baxter


  • NDMS does not include all hospitals, but includes the majority.
  • Show NDMS to another level—go to other system.
  • "Throughput" is the limiting factor for the ultimate number of available beds.
  • Tracking hospital resources to calculate throughput.
  • HAvBED does not have throughput capabilities; hospitals do not have technology to make this judgement.


  • Need to have hospitals regulate operations during tests of HAvBED.
  • Set conference call for the April meeting.
  • Change the date for the April meeting? NDMS conference.

Next Meeting Date

HAvBED project team will have a phone conference sometime in April for the next meeting and will solicit votes.

HAvBED Advisory Group Meeting Minutes
Thursday, October 14, 2004
Monday, May 2, 2005

Hotel Caribe Royale
Orlando, Florida


Lt. Col. James Baxter
Mark Roupas
Sally Phillips, R.N., Ph.D.
Stephen Cantrill, M.D.
Duane Mariotti
Jacob Dye
LCDR Cipriano Pineda
Capt. F. Christy Music, M.Sc., U.S.N.
Shayne Brannman
Chris Felton
Andrew Nunemaker

PowerPoint® Presentation

Dr. Cantrill presented the HAvBED project progress leading up to data delivery by system partners and hospitals using the manual data entry Web interface:

  • Secured Server online.
  • Database ready to receive data.
  • System partners testing Web services interface.
  • GIS interface progress.
  • Requested matrices for final report.

Data Points

EMSystem®, represented by Andy Nunemaker and Chris Felton, explained how data is requested from participating hospitals it covers.  A data entry screen is dispersed to the data entry person at each participating hospital as an HTML page.  Data is entered and sent to an EMSystem® database and on to the HAvBED database.  Hospitals participating through EMSystem® will include all the public and private hospitals in the State of Colorado, and potentially hospitals in Southeast Wisconsin and Virginia.  State Hospital Capacity System, represented by Duane Mariotti, explained that a subset of data already being requested from participating hospitals in Washington, Oregon, and South Carolina will be sent to the HAvBED database.  Hospital data will include public, private, and military hospitals. 

Matrices for Final Report

Dr. Cantrill requested the advisory group discuss the final report documentation and the project findings in the form of matrices. 

  • The one suggestion that dominated the discussion was the need for a followup survey to determine the success of the HAvBED project from the user's perspective.  Many of the obstacles the HAvBED project has met were not technological but administrative.  Interfacing with hospital data, which is considered proprietary to the individual hospital, presents many challenges.  To ensure the continuance of the HAvBED project requires the documentation of the process to construct the hospital database, GIS and Web interface.  In addition, the process each individual hospital went through administratively to approve or disapprove the participation in the HAvBED pilot project will assist in a full-scale national implementation.  The survey would document both the technological and administrative issues involved with participating in the HAvBED project.
  • It was suggested that HAvBED staff file for an extension through AHRQ for the followup survey and documentation.

HAvBED Advisory Group Meeting Minutes
Thursday, Tuesday, July 12, 2005

Deputy Secretary's Conference Room
Office of the Assistant Secretary for Public Health Emergency Preparedness
Hubert Humphrey Building, 200 Independence Ave., Rm. 305A
Washington, D.C.  20201


Sally Phillips, R.N., Ph.D.
Robert Claypool, M.D.
Padmini Jagadish
Stephen Cantrill, M.D.
Nathaniel Hupert, M.D., MPH
Peter Pons, M.D.
Jacob Dye, M.S.
Duane Mariotti
Daniel Salazar
Sheri Eisert, Ph.D.
Andrew Nunemaker
Michael Feeser
Kevin Yeskey, M.D.
Capt. F. Christy Music, M.Sc., U.S.N.


HAvBED Presentation: Dr. Stephen Cantrill

Dr. Cantrill reviewed the steps involved in the development of the HAvBED project including the development of bed definitions, the designation of desired data elements and the development of the operational HAvBED system including the specification of XML interface, the Emergency Data Exchange Language.  A review of the three testing periods was presented.  The functionality of the HAvBED system was demonstrated via an internet connection to the HAvBED system.  Operational issues and limitations were presented.  Recommendations for next steps were discussed.  Advisory group members were provided with a logon and password for the HAvBED system.  They were encouraged to review the system and provide any comments to Dr. Cantrill.


  • In general, the Advisory Group thought the HavBed System is a good tool that could be used by regional and local emergency planners and responders.
  • Edward Gabriel expressed that the HavBEd system would be good tool for emergency planners, especially when dealing with interState issues of bed availability.
  • There were discussions of next steps concerning ownership and sustainability of the HAvBED system. Dr. Claypool and Phillips expressed the need to have a followup meeting regarding how to proceed once the Final Report is completed and recommendations finalized.

Return to Contents
Proceed to Next Section

The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care